Why We Hate the Diamondbacks

A year ago, the Arizona Diamondbacks went 79-83. Over the winter, they signed Zack Greinke and Tyler Clippard as free agents, plus they notably traded for Shelby Miller and Jean Segura. Reinforced with one of the game’s best pitchers, a quality starter, a good reliever, and a middle infielder with a pulse, we currently have the 2016 Diamondbacks projected to go… 79-83. And not surprisingly, Arizona’s GM doesn’t think we’re going to be right on this one.

Q: Does that make any sense to you? You add Greinke, you add Miller, and the math boys say you are not going to win any more games?

Stewart: “Jack, I think out there there are a lot of people that don’t want us to win. For those people that don’t want us to win, that’s OK. We’re still going to play the game the same way. We embrace the challenge every day of coming out and playing and doing the things that we’re capable of doing. And those who think that we’re a 78-win team, you know what? That’s what they think. When you start making predictions like that and you keep coming up wrong, you lose credibility.”

Q: Why do you think there are people who want you to lose?

Stewart: “I think the way that we do things. We’re a baseball team here. We believe in our team and how we play the game. I just think, in everything, there is always everyone who doesn’t want to see you do well. Obviously, anybody who says we can only win 78 games, they’re either not thinking or they’re not believing that what we have here is a team that’s capable of winning more games than that. So when I say that there are people out there who do not want us to win, that’s a prime example of that. To think we will only win 78 games? That’s a joke.”

Q: Do you think they are taking a shot at the old school, fundamental approach?

Stewart: “I try not to even think with people like that. I try to think with the people who think logically. And if you are thinking logically and we won 79 games last year, with the additions of Greinke, Miller, Clippard, Segura, people that make your team better, I think it is impossible for us to only win 78 games. Like I said, they predicted we would lose 96 last year.”

(For the record, we actually projected that the D’Backs would lose 89 games a year ago, and BP had them as an 88-loss team, so Stewart’s information is a bit off, but his overall point that they played better than we expected a year ago is indeed correct.)

Now, we’ve certainly leveled a fair number of criticisms towards the Diamondbacks over the last couple of years, so Stewart’s belief that we’re against him or his organization is understandable, at least if we try to see it from his perspective. Discrediting your critics is a pretty natural human response, and I don’t think it’s particularly surprising that he wouldn’t care what we have to say about his team’s chances in 2016. If I was in his position, I’d likely have the exact same reaction.

But I figure it is probably worth going through the exercise of seeing why our forecasts don’t think the Diamondbacks are going to be significantly better than they were a year ago, despite their big offseason moves. After all, despite what some might think, we really don’t tweak the algorithm to produce results that we find personally satisfying; beyond the manually-curated depth charts that provide the playing time forecasts, we have no ability to manipulate the projections. ZIPS, Steamer, and PECOTA are non-emotional lines of code, not the personal beliefs of any individual human being with rooting interests.

So why don’t the numbers think the 2016 D’Backs are going to be significantly better than the 2015 version? Let’s look at the team’s position players from a year ago, and how their projections stack up for 2016.

2015 vs 2016 Diamondbacks Hitters

Name	2015 WAR	2016 WAR	Difference
Paul Goldschmidt	7.4	5.7	-1.7
A.J. Pollock	6.6	3.9	-2.7
David Peralta	3.7	1.8	-1.9
Ender Inciarte	3.3	0.0	-3.3
Jake Lamb	2.0	2.2	0.2
Welington Castillo	1.7	1.5	-0.2
Nick Ahmed	1.7	0.8	-0.9
Jarrod Saltalamacchia	1.1	0.0	-1.1
Mark Trumbo	0.7	0.0	-0.7
Phil Gosselin	0.5	0.2	-0.3
Socrates Brito	0.4	-0.1	-0.5
Peter O’Brien	0.2	0.5	0.3
Aaron Hill	0.1	0.0	-0.1
Jamie Romak	0.1	0.0	-0.1
Gerald Laird	0.0	0.0	0.0
Jordan Pacheco	-0.1	0.0	0.1
Oscar Hernandez	-0.1	0.0	0.1
Tuffy Gosewisch	-0.1	0.0	0.1
Brandon Drury	-0.2	0.1	0.3
Daniel Dorn	-0.2	0.0	0.2
Cliff Pennington	-0.3	0.0	0.3
Yasmany Tomas	-1.3	-0.1	1.2
Chris Owings	-1.4	0.0	1.4

You can probably see a primary source of disagreement right at the top of the table. A year ago, Goldschmidt and Pollock were perhaps the best pair of teammates in the game, combining for +14 WAR between them. While both are excellent players and are expected to perform at high levels again, their 2016 projection has them combining for +10 WAR, a pretty significant step down from what they did in 2015. And this isn’t because Steamer just hates Arizona; history shows that basically any player who has the best year of his career should be expected to perform worse the next year. The same forecasting system has Bryce Harper going from +9.5 to +6.6 WAR, for instance. Josh Donaldson is expected to go from +8.7 to +6.1. It’s not that Steamer is down on Goldschmidt or Pollock; it’s just that they set such a high bar that it’s unreasonable to expect them to repeat it.

Then, we get into the two corner outfield spots, which is probably where the most disagreement between the forecasts and the D’Backs perception lies. Because of David Peralta’s weird career path, the projections aren’t particularly high on him, and have him losing half of his value from a year ago. Additionally, Ender Inciarte’s +3 WAR season can’t be repeated because Ender Inciarte isn’t on the team anymore, having been traded to Atlanta in the Miller deal. The projections expect Yasmany Tomas to be a lot better than he was a year ago, but that simply moves him from outright disaster to guy-who-still-shouldn’t-be-playing, and so instead of getting +7 WAR from Peralta and Inciarte, the projections have the D’Backs getting just +2 WAR from Peralta and Tomas.

If you’re going to quibble with the forecast, this is probably your best bet. Peralta’s path to the majors makes him particularly difficult to find comparisons for, and the system may be overcompensating for the fact that he got to the big leagues at a late age. Personally, I’d take the over on Peralta’s projection for 2016, and think he’s probably going to be a pretty good hitter again next year. Tomas, I’m less sold on, but he was a pretty highly thought of international signing just a year ago, so this is another player where the projections are dealing with larger error bars due to more limited performance data. If you think Tomas can be a reasonably productive regular — and the Diamondbacks basically have to think that, given that they traded Inciarte away to open up a job for him — then you can mentally add a couple of wins to account for that optimism.

The rest of the hitters don’t move the needle much. Jean Segura’s +1 WAR projection isn’t accounted for in the table above, but his acquisition and the dead-cat-bounce that Chris Owings gets in his forecast are mostly offset by the reduced production from Nick Ahmed and the loss in value from not having Jarrod Saltalamacchia’s 200 good plate appearances again. By and large, the projections see mostly the same production out of the role players, so the position players drop off from +25 WAR (#6 in MLB in 2015) to a projected +18 WAR (#22 in MLB for 2016). While it’s primarily the same group of players, regression from two guys having career years, the loss of Inciarte, and the expected step back from Peralta cost the team seven wins off of last year’s production.

Of course, the D’Backs would counter that their big upgrades were on the pitching side of things, and losing Inciarte got them Shelby Miller, plus Greinke and Clippard only cost them money, and now they’ll have Patrick Corbin for a full season. The pitching has to be better, right?

Absolutely. A year ago, the D’Backs pitching staff ranked 27th in the majors with just +8 WAR, and they will almost certainly be a lot better than that this year.

But even with all the changes, this still isn’t going to be a great pitching staff. Greinke is excellent, and you can make a pretty decent case that Steamer is underselling Miller’s expected production given that he’s almost 600 innings into a career where he’s significantly outperformed his peripherals. But if you decide to reject Steamer’s pitching projections in favor of looking at career ERA to this point, the negative adjustment for Rubby de la Rosa is probably larger than the positive adjustment for Miller, since Rubby has underperformed his peripherals by a good margin.

And then there’s the issue of depth. Corbin, de la Rosa, and Ray are all significant question marks, but if any of them falter, the team’s collection of potential replacements are all pretty close to replacement level. This is a rotation that could be pretty solid if the primary five guys each make 30 starts, but things fall apart quickly if someone gets hurt. And pitchers get hurt. The lack of depth is one of the main reasons why the D’Backs rotation projects for +12.5 WAR, putting them in the middle of the pack in terms of starting rotations for 2016. That’s six wins better than the rotation put up a year ago, so the rotation upgrades wipe out almost all of the deficit we started with on the position player side of things, but it doesn’t push them forward enough.

In the bullpen, things are projected to be mostly the same. Last year, the D’Backs got +2 WAR based on a 3.91 FIP, but they significantly outperformed that as a group, and posted a 3.50 ERA. This year, Arizona’s relievers are projected for +2 WAR based on a 3.82 FIP, and again are expected to beat that handily, putting up a 3.50 ERA due to having guys like Clippard who have shown that they can generate weak contact. In the bullpen, the names have changed, but the expected overall performance is basically the same.

So, the short summary is simply that the upgrades to the pitching staff look to account for around the same margin as the expected downgrades on offense and defense. If you think that Goldschmidt and Pollock can repeat their +7 WAR seasons, then add a few more wins to the ledger. If you think Peralta is going to build on his breakout season rather than take a step back, add a win or two. If you think Tomas is going to justify the hype and prove he can play in the big leagues, then add a few wins.

And if you make all of those assumptions, then yeah, the D’Backs look like a team that should win something like 85-95 games, depending on health and what kind of midseason acquisitions they make to bolster themselves for a playoff run. But let’s be honest; we could do that kind of best-case-scenario wishcasting for almost any team in baseball. If everyone’s best players repeat their career years, and all the guys who sucked last year get a lot better, and no one gets hurt, then every team in baseball is a legitimate contender.

In reality, guys who play at elite levels usually get worse, even if they’re still really good. Pitchers get hurt, and sometimes, guys who are terrible stay terrible. There is absolutely a scenario where the Diamondbacks outperform these projections, and I’d probably take the over on enough of Steamer’s projections to see them as an 80 or 81 win team myself, but to get to the point where “78 wins is impossible”, one has to take an extraordinarily optimistic view of almost every player on the roster.

The forecasting systems aren’t trying to project best-case outcomes. Instead, they simply look at a Diamondbacks roster that has some key players who are likely to perform worse than they did a year ago, sees zero depth in the rotation, and several significant question marks among the projected regulars, without any obvious internal replacements around if things go south again. It’s a team that absolutely could win if enough things go right, but while the new guys they got will help, the team isn’t adding those guys to a 79-win base. That’s just not how baseball works.